Suicide: a Dangerous Undertow in the
Lives
of Patients With Bipolar Disorder
By Arline Kaplan © 1999 (All Rights Reserved)
Over the years, my manic-depressive illness
became much worse, and the reality of dying young from suicide became a dangerous
undertow in my dealings with life. Then, when I was twenty-eight years old,
after a damaging and psychotic mania, followed by a particularly prolonged and
violent siege of depression, I took a massive overdose of lithium. I unambivalently
wanted to die and nearly did. Death from suicide had become a possibility, if
not a probability, in my life. —Kay Redfield Jamison, Night
Falls Fast
Paradoxically and yet fortuitously, at the same time Kay Redfield Jamison, Ph.D.,
was experiencing this "mood of death," she was a young faculty member
in an academic psychiatry department. She sought to study everything she could
about her disease and the psychological and biological determinants of suicide.
Now, a professor of psychiatry at The Johns Hopkins School of Medicine and an
internationally recognized authority on manic-depressive illness, Jamison has
written, Night Falls Fast—Understanding Suicide, a book about
the psychology of suicide and suicide as a medical and social problem.
Jamison is not alone in her efforts to raise public and professional awareness
about suicide and its links to psychiatric illness. Psychiatrists Sylvia Simpson,
M.D., and Jan Fawcett, M.D., among many others are speaking at national conferences,
writing journal articles and informing advocacy groups. In recent interviews
Jamison, Simpson and Fawcett shared their concerns and expertise.
"Between 15% and 20% of people with severe, untreated bipolar illness will
die by suicide, and between 25% and 50% of people with bipolar illness attempt
suicide at least once [Goodwin and Jamison, 1990]," Jamison said. In comparison
to the general population, individuals with bipolar disorder are 15 times more
likely to die by their own hand (Harris and Barraclough, 1997; Simpson and Jamison,
1999).
At high risk for suicide are those early in the course of illness, Jamison said.
"After their first episode is a particularly high-risk time."
That tendency surprises many people, said Simpson, associate professor of psychiatry
at Johns Hopkins. Some psychiatrists, she said, speculate that because first-episode
patients haven’t had the experience of recovering from an episode, they
"feel it is never going to end and [they] can’t imaging going on feeling
like that."
"Patients who do well socially and academically when young and who then
are hit by devastating illnesses such as schizophrenia or manic-depression seem
particularly vulnerable to the spectre of their own mental disintegration and
the terror of becoming a chronic patient," wrote Jamison in Night Falls
Fast. "For them and many others, there is a terrible loss of dreams
and inescapable damage to friends, family and self."
Other high risk factors cited by the experts include comorbidity, access to
firearms, especially violent suicide attempts, family history of suicidal behaviors,
certain age groups, depressed or mixed states, and symptoms such as anxiety
or impulsivity.
Comorbidity is very important, Simpson said, particularly comorbid substance
abuse but also anxiety disorders and personality disorders.
"The more comorbid conditions, the higher the suicide risk," she said.
Two out of every three people with manic-depression have substantial alcohol
or drug problems, Jamison wrote in her book, noting, "Substance abuse loads
the cylinder with more bullets by acting to disinhibit behavior. Drugs and alcohol
increase risk taking, violence and impulsivity. For those who are suicidal or
potentially so, this may be lethal." (In an article by Brady and Lydiard
(1992), the prevalence of substance abuse among patients with bipolar affective
disorder ranged from 21% to 58%—Ed.)
Simpson was also concerned about the availability of firearms and the increasing
tendency among women with bipolar disorder to use more violent means to kill
themselves.
She described a study by Wintemute and colleagues (1999) that found suicide
was the leading cause of death among handgun purchasers in the first year after
purchase, accounting for 24.5% of all deaths and 51.9% of deaths among women
aged 21 to 44 years. In the first week after purchase, the rate of suicide by
means of firearms among purchasers was 57 times as high as the adjusted rate
in the general population.
Women who have bipolar disorder most often overdose on medication (usually antidepressants,
combined with tranquilizers and alcohol), Simpson said, but many of them are
using more lethal means, including guns, hanging and jumping.
While suicide deaths from bipolar disorder are usually assumed to be associated
with the depressed phase of the illness, Jamison said the "mixed states
are probably the single most lethal state, because you have all of the worst
aspects of depression along with a terrible energy, perturbation and agitation."
Fawcett, professor and chairman of the department of psychiatry at Rush-Presbyterian-St.
Luke’s Medical Center in Chicago, noted that 40% to 45% of bipolar patients
go into mixed states.
"In mixed states, they are frequently agitated, hyperactive, impulsive,
angry and irritable and at the same time, depressed," Fawcett said. "It
is extremely painful to be in mixed states. Patients are suffering tremendously
and because they are depressed, they don’t see it resolving. It is like
being tortured with no escape, so suicide becomes the way out."
Short-term Risk Factors
Frequently symptoms, such as anxiety and agitation, can serve as signals for
potential suicide in bipolar and unipolar patients, said Fawcett who is also
director of the Rush Institute for Mental Well-Being.
As part of the Collaborative Study on the Psychobiology of Depression funded
by the National Institute of Mental Health, Fawcett and colleagues (1990) compared
prospective clinical data in 32 patients who committed suicide with 922 who
did not. All the patients were diagnosed as having a major affective disorder.
Fawcett and his colleagues found two types of risk indicators for suicide.
The typical ones, he said, included suicidal ideation, suicide plans, suicidal
impulses expressed to another person, a history of suicide attempts. The presence
of psychosis was also of importance in some of the patients.
"[These] are things that are supposed to be assessed in every patient when
they have depression or manic-depressive illness. What we found is those indicators
predicted suicide in two to 10 years from the time we interviewed the patient,
but they didn’t predict suicide in the first weeks or months or year from
the interview," he said.
A primary indicator of acute suicidal risk was severe psychic anxiety as demonstrated
both clinically and by elevated 24-hour 17-hydroxycorticosteroid (Fawcett, 1988).
Fawcett characterized psychic anxiety as "constant worry or fear about
everyday things or about having made a horrible mistake or broken the law or
financial pressures…that’s all the person can think about."
Other indicators associated with suicide within one year of the initial assessment
included panic attacks, anhedonia (severe loss of interest or pleasure), diminished
concentration, global insomnia and moderate alcohol abuse.
Similar indicators of acute suicide risk emerged in a subsequent study of 14
inpatient suicide charts from university, community and state hospitals across
the nation, using 24-hour staff notes recorded within one week of suicide (Busch
et al., 1993).
Thirteen patients (93%) showed definite evidence of psychic anxiety. The study
investigators concluded, "comorbidity of moderate to severe psychic anxiety
with other psychiatric diagnoses, particularly in the context of psychosis and/or
depression, may increase the risk of suicide."
That study of inpatient suicides has since been expanded to 100 cases. Fawcett
said he and colleagues have analyzed data on 76 cases.
In the expanded study, chart records showed that 79% of the patients exhibited
severe agitation and anxiety. Other symptoms included panic attacks, global
insomnia, inability to concentrate and anhedonia. Those symptoms were recorded
by hospital personnel who were unaware patients were going to commit suicide
a week later, Fawcett emphasized.
"Subsequent to our studies, Hall and others (1999) reviewed risk factors
for suicide in 100 patients who made severe suicide attempts…they found
a high level of anxiety and panic in these patients. Their findings are important
to me because it is the first replication outside of our collaborative data
and our own data."
Treatment
In his treatment of bipolar patients, Fawcett focuses on the "behavioral
dimensions"—anxiety, agitation, impulsivity, depression, psychosis—that
are associated with increased risk of suicide.
"Looking for dimensions we can treat to prevent suicide is what I am excited
about…not giving up diagnostic systems, but adding this dimensional approach
in terms of treatment decisions is important," he said.
"You can treat anxiety and agitation much more rapidly than you can depression
which takes sometimes weeks to respond to treatment," he added. "So
if clinicians recognize anxiety and treat it, they can actually lower the risk
of death in patients while treating the depression."
To "settle down the agitation and anxiety," Fawcett recommended using
high-dose benzodiazepines such as lorazepam (Ativan) or alprazolam (Xanax) for
immediate treatment. Certain borderline patients, who may become disinhibited
by benzodiazepines, are exceptions.
"For longer-term treatment, you can use medications such as the anticonvulsant
divalproex (Depakote)," he said, adding that divalproex also works well
for agitation. Other medications he mentioned included atypical antipsychotics,
such as olanzapine (Zyprexa) and sedating neuroleptics, such as thioridazine
(Mellaril) as treatment options for the agitation and anxiety.
In her practice, Simpson said, she sees moderately to severely depressed patients
on a frequent and regular basis and always asks about their suicidal thoughts.
"Often the depressed mood, the hopelessness and the suicidal thoughts are
among the last symptoms of depression to improve," she said. "The
person may be eating and sleeping better and may have more energy, but if they
are still hopeless and suicidal, they may be at increased risk of committing
suicide."
To try to assess the acuity of suicide risk, Simpson uses a hierarchy of questions.
She will ask patients what the suicidal thoughts are; how often the thoughts
are occurring; how much time they spend thinking about suicide; whether they
are able to resist such thoughts; whether they have a plan and have made any
preparations toward acting on the plan; whether they have said their good-byes
or gotten their affairs in order; and whether they have access to lethal means,
such as guns.
"I always ask people how close they have come to acting on their suicidal
thoughts and how likely they think they are to act on them," she said.
"Some people won’t tell you what they are thinking and planning, but
others will. It is really important to learn as much as you can and then make
the decision as to what kind of care the person needs at the moment."
While careful questioning is important, Fawcett warned that in his experience
and research many patients who commit suicide often deny their suicidal intent.
In the prospective study, Fawcett and colleagues (1990) found that most of the
patients who completed suicide in the first year after assessment did not communicate
the presence of suicidal ideation or plans in response to the specific questions
of trained, clinically experienced raters.
In the expanded inpatient study, Fawcett said that with 77% of the patients
who killed themselves, there was a note in their charts in which they clearly
denied suicidal intent.
Overall, Jamison said, physicians need to be better trained to make the differential
diagnosis between bipolar II and major depression and to pick out the subtle
signs of bipolarity.
"They need to ask more questions about mild, manic states, and they need
to ask more questions about irritability, volatility and impulsivity,"
she said. "[Mental health professionals] need to be more aware of the dangers
of mixed states, and they need to be more aware of the literature showing that
lithium (Eskalith, Lithobid) has a strong antisuicide effect [Müller-Oerlinghausen
et al., 1992]."
Discussing anticonvulsants, Simpson said, "Intuitively, one would think
that since they act as mood stabilizers, decreasing cycling into depression
and mixed states that they would be useful in decreasing suicides." (Charles
Bowden, M.D., of the University of Texas Health Science Center in San Antonio,
has presented some data on the effects of divalproex and lithium and suicidality
in bipolar patients—Ed.)
Coupled with improved training for physicians and other clinicians, Jamison
added, "we need more depression awareness and screening programs in the
schools, and parents need to be more educated about the symptoms of depression."
The Surgeon General’s Call to Action To Prevent Suicide (Satcher, 1999),
Jamison said, provides a good national plan. It includes recommendations to
broaden the public’s awareness of suicide and its risk factors; to enhance
services and programs, both population-based and clinical care; and to advance
the science of suicide prevention.
Research
Because both Simpson and Fawcett have been actively involved in research on
suicide prevention and treatment, they were asked about their current work.
Simpson explained that the team headed by J. Raymond DePaulo Jr, M.D., at Johns
Hopkins has been conducting family and genetic studies relevant to bipolar disorder
since 1986. In the last two years, Simpson and others have been doing follow-ups
on members of bipolar study families who had reported making a suicide attempt.
"We asked them about subsequent attempts and tried to learn more about
their most severe attempts. We asked whether they were in a depressed or mixed
state at the time of the attempt, and whether they were drinking or using drugs
at the time. We asked how severe the attempts were and what kind of treatment
they required afterward. We also asked about any aggressive behaviors towards
other, which is our study families appeared to be infrequent."
Meanwhile the genetic linkage studies of bipolar disorder are continuing.
"Our group and others have identified a region on chromosome 18 that may
be associated with bipolar disorder in a subset of families," she said.
"We are trying to narrow down the region on chromosome 18 in hopes of finding
a gene or genes for bipolar disorder. We would also look to see if that locus
is associated with increased susceptibility to suicidal behaviors
As part of their research efforts, Fawcett and his team (Katie Busch, M.D.;
Louis Fogg, Ph.D.; and Patricia Meaden, Ph.D., are developing an assessment
to assess the presence of anxiety, worry and other factors.
"We are getting ready to test it on an inpatient sample to see how it distinguishes
in terms of the patient’s history and other critical information. It is
hard to prove you can prevent suicide, he said.
Fawcett is also interested in the neurobiology of suicide. In a chapter, "Suicide:
A Four-Pathway Clinical-Biochemical Model," Fawcett and colleagues (1997)
looked at four hypothetical pathways leading to suicide in clinical depression.
The first pathway involved links between severe anxiety/agitation and high brain
corticotrophin-releasing factor (CRF ). The second looked at baseline hopelessness
and reactivity hopelessness (defined as the rate of increase in hopelessness
as a function of depression symptom severity) as character traits related to
chronic suicide risk. The third looked at anhedonia severity as a risk factor
and character trait for chronic suicide risk. The fourth pathway examined the
relationship between low cholesterol and low brain serotonin function and impulsive/aggressive
behavior.
"These are hypotheses, but there is some evidence…to support [them],"
Fawcett said. For example, high CRF has been found in the brains of suicide
victims. Evidence supports a mechanism linking increased corticotrophin-releasing
hormone, at least hypothetically, to severe anxiety/agitation in the depressed
patient with evidence of HPA (hypothalamic-pituitary-adrenal) axis hyperactivity
(Fawcett et al., 1997).
There is already evidence in animal studies that alprazolam (Owens et al., 1993)
and valproate block corticotropin-releasing factor in stressed animals (Nemeroff
et al., personal communication), Fawcett said.
On the question of low cholesterol, Fawcett and colleagues acknowledged that
some studies have not shown an association between low cholesterol and suicidal
behavior. However, they then presented data on 49 cases of inpatient suicide
with recorded serum cholesterol levels. The cases were compared to the age and
sex-corrected levels of the National Lipid Study (Fawcett et al., 1997).
"We found significantly lower levels of cholesterol among the patients
who committed suicide compared to national samples," he said. A lot of
data exist "on cholesterol and suicide attempts in large studies, but there
has not been many studies of actual suicides where the patient was actually
in treatment."
References
Brady KT, Lydiard RB (1992), Bipolar affective disorder and substance abuse.
J Clin Psychopharmacol 12(Suppl):17S-22S.
Bursch KA, Clark DC, Fawcett J, Kravitz HM (1993), Clinical features of inpatient
suicide. Psychiatric Annals 23(5):256-262.
Fawcett J (1988), Predictors of early suicide: identification and appropriate
intervention. J Clin Psychiatry 49(10 suppl): 7-8.
Fawcett J, Busch KA, Jacobs D et al. (1997), Suicide: a four-pathway clinical-biochemical
model. In: The Neurobiology of Suicide From the Bench to the Clinic, Stoff DM,
Mann JJ, eds. New York: Annals of the New York Academy of Sciences, Vol. 836.
Fawcett J, Scheftner WA, Clark DC et al. (1987) Clinical predictors of suicide
in patients with major affective disorders. A controlled, prospective study.
Am J Psychiatry 144:36-40.
Fawcett J, Scheftner WA, Fogg L et al. (1990), Time-related predictors of suicide
in major affective disorder. Am J Psychiatry 147(9):1189-1194.)
Goodwin FK, Jamison KR (1990), Manic-Depressive Illness. New York: Oxford University
Press, pp. 227-244.
Hall RC, Platt DE, Hall RC (1999), Suicide risk assessment: a review of risk
factors for suicide in 100 patients who made severe suicide attempts. Evaluation
of suicide risk in a time of managed care. Psychosomatics 40(1):18-27.
Harris EC, Barraclough B (1997), Suicide as an outcome for mental disorders.
A meta-analysis. Br J Psychiatry 170:205-228.
Jamison KR (1999), Night Falls Fast—Understanding Suicide. New York: Alfred
A. Knopf.
Müller-Oerlinghausen B, Muser-Causemann B, Volk J (1992), Suicides and
parasuicides in a high-risk population on and off lithium long-term medication.
J Affect Disord 25:261-270.
Owens MJ, Vargas MA, Nemeroff CB (1993), The effects of alprazolam on corticotropin-releasing
factor neurons in the rate brain: implications for a role for CRF in the pathogenesis
of anxiety disorders. J Psychiatr Res 27(Suppl 1):209-220.
Satcher D (1999), The Surgeon General’s Call To Action To Prevent Suicide.
Accessed at http://www.surgeongeneral.gov/library/call to action/default.htm.
Downloaded Dec. 9, 1999.
Simpson SG, Jamison KR (1999), The risk of suicide in patients with bipolar
disorders. J Clin Psychiatry 60(suppl2):53-56.
Wintemute CJ, Parham CA, Beaumont JJ et al. (1999), Mortality among recent purchasers
of handguns. N Engl J Med 341(21):1583-1589.
Viewers, Your Help Is Needed:
I am a freelance medical journalist. Some of the articles posted on this Web site were originally written for health care professionals, but I believe patients/consumers may also find value in the articles. Many of us who have chronic or life-threatening health problems become experts on them. For that reason, I have kept the rights to electronically post these articles.
Additionally, I am developing new stories specifically for this site. It is my intention to offer you high-quality, well-researched information. Because this site accepts no advertising, your financial support is needed to continue this user-sponsored service.
I welcome your comments, suggestions and financial help. You can send voluntary contributions to, or contact me at:
HealthRising
Arline Kaplan
P.O. Box 3644
Costa Mesa , CA 92628